Comparative Analysis of Machine Learning and Neural Network Models for Wine Quality Prediction

Abstract

The assessment of wine quality is of paramount importance to both consumers and the wine industry. Recognizing the evaluation of wine quality holds great importance for both consumers and the wine industry. As awareness of its influence on customer satisfaction and business success grows, companies are increasingly turning to product quality certification to bolster sales in the global beverage market. This certification process plays a vital role in enhancing the market value of wine products. In the past, quality testing typically occurred towards the end of the manufacturing process, leading to time-consuming and resource-intensive procedures. This conventional approach involved the participation of numerous human experts tasked with assessing wine quality, resulting in significant costs. Additionally, since taste perception is subjective and varies among individuals, relying solely on human specialists for wine quality assessment presents notable challenges. To tackle these obstacles, our research aims to advance the field of wine quality prediction by harnessing the diverse characteristics of wine. In our study, we utilized three distinct datasets to train and test our predictive models. We employed various feature selection techniques and explored machine learning algorithms such as XGBoost, Random Forest, Naive Bayes, and Neural Networks to identify the optimal combination of parameters for accurate wine quality prediction. By transitioning from subjective human evaluation to data-driven approaches, we seek to enhance the efficiency and reliability of wine quality assessment. Our research represents a significant stride forward in the field, enabling the prediction of wine quality based on objective factors. This approach not only reduces the time and costs associated with traditional quality assessment methods but also establish a more standardized and consistent evaluation process. Ultimately, our findings contribute to the advancement of wine industry practices, empowering businesses to make informed decisions and deliver high-quality products that meet consumer expectations.